Cookbook

This section is a companion to the Pixano Cookbook repository. Each chapter takes a well-known computer vision benchmark, explains the problem it addresses, shows how its data structure maps to Pixano's data model, and walks you through the import and annotation workflow. Along the way, you will discover the Pixano features that are most relevant to each task.

🎯

Object Detection

Learn how Pascal VOC maps to Pixano's data model. Import images, annotate with bounding boxes, and explore interactive segmentation.

Video Object Tracking

Learn how DAVIS maps to Pixano's video model. Understand tracklets, per-frame masks, and temporal annotation workflows.

Multi-View Detection

Import FLIR ADAS v2 in image or video mode, and browse synchronized RGB + thermal views side by side.

Visual Question Answering

Learn how VQAv2 maps to Pixano's data model. Combine Q&A annotation with spatial entity annotation in the same workspace.

Entity Linking

Import the MEL dataset, link image regions to text spans, and annotate with bounding boxes and text selection.